JP2019525320A - Apparatus for assessing medical device quality - Google Patents

Abstract

The present invention relates to an apparatus for assessing medical device quality. Providing (210) at least one report associated with at least one medical image collected by the medical device is provided, wherein the report is associated with a corresponding medical image. At least one report is analyzed and at least one image quality parameter is generated (220), and the image quality parameter is associated with the corresponding report. At least one image quality parameter is evaluated and alert information related to the medical device is generated (230).

Description

  The present invention relates to an apparatus for assessing medical device quality, a system for providing medical device alerts, a method for assessing medical device quality, a computer program element and a computer readable medium.

  The general background of the present invention is the field of determining the quality of medical devices. The cost effective use of medical imaging modalities requires high system availability. Generally, a hospital has a contract with a maintenance service provider. Most maintenance is planned maintenance or collective maintenance. In the latter case, when a system malfunction is detected, a call is made to the service center and the service engineer starts troubleshooting.

  US 2013/0251219 A1 describes that a medical image quality reporting and monitoring system for use in a medical imaging system comprises a medical image computer comprising a display processor, a display and a report generator. The display processor generates data representing an image for display that includes a user-selectable image element that allows the user to identify the at least one medical image as having an image quality defect. The display presents an image. The report generator automatically generates a report in response to detecting a selection of image elements that identify at least one reduced quality medical image as having an image quality defect. The report includes data representing an anonymized quality-reduced image having image quality defects, the time of collection of the quality-reduced image, and the imaging system collection settings used in collecting the quality-reduced image.

  It would be advantageous to have an improved apparatus for assessing medical device quality and an improved system for providing medical device alerts using such apparatus.

  The object of the present invention is solved using the subject matter of the independent claims and further embodiments are incorporated in the dependent claims. The following described aspects and examples of the invention include an apparatus for assessing medical device quality, a system for providing medical device alerts, a method for assessing medical device quality, a computer program element, and a computer readable medium. Note that also applies.

According to the first aspect,
-An input unit;
An apparatus for assessing medical device quality comprising a processing unit is provided.

  The input unit is configured to provide the processing unit with at least one report associated with at least one medical image collected by the medical device, the report being associated with the corresponding medical image. The processing unit is configured to analyze the at least one report and implement a classifier module for generating at least one image quality parameter, the image quality parameter being associated with the corresponding report. The classifier module is configured to apply a natural language algorithm to the at least one report to generate at least one image quality parameter. The natural language algorithm includes a learning algorithm, and the learning algorithm is configured to generate a value for the image quality parameter based on the at least one training data. The processing unit is also configured to implement an evaluation module for evaluating at least one image quality parameter and generating alert information related to the medical device.

  In this way, image quality parameters are derived from reports associated with images collected by the medical device, which are used to generate alert information. This allows predictive maintenance of the medical device to be performed and is used as background information for the device when routine scheduled maintenance is performed, e.g. when a call is made to a service center It can also help with troubleshooting.

  In other words, the natural language as a word used by a radiologist, for example when preparing a report on a medical image, is analyzed, thereby enabling effective use of the report to determine information related to the medical device. Can provide a way.

  In other words, the learning algorithm is trained (supervised) using reports (such as radiology reports) where text fragments related to image quality issues may be manually labeled as such. Learning).

  In this way, the accuracy of image quality parameter generation can be continually improved as more reports are analyzed. For example, such labeling by periodic manual labeling of the reports described above, verifying that there was such an image problem identified by a radiologist, is a technique familiar to imaging technicians or medical devices. The learning algorithm is improved through operations on such “supervised” input data.

  Data protection issues related to patient image analysis can be problematic even when processing to anonymize the image is performed, and image display and processing requires significant computational resources and is difficult It can be. However, by deriving image quality parameters from reports associated with images collected by the medical device, such data protection of image analysis is not an issue and processing is made more efficient and simple.

  Thus, in this way, the apparatus automatically detects when an image defect has been detected. This is less disturbing for the radiologist so the radiologist can continue to evaluate the image.

  In this way, information related to the image quality provided by the medical device is extracted to help evaluate the performance of the medical device.

  This allows the behavior of the medical device to be determined based on the report without the need for extensive image quality analysis of the image to identify artifacts in the image that indicate a problem with the medical device.

  In addition, the assessment of the workings of medical devices based on reports means that extensive analysis of the image is not required to identify device problems, where such unrelated analysis of the image is Can be problematic due to data protection issues.

  The apparatus has the advantage that medical device quality can be evaluated without the need for image processing to take place. Also, image quality parameters from the medical device are not needed to evaluate the quality of the medical device.

  In one example, the report of the at least one report includes text, and the classifier module is configured to analyze at least one portion of the text to generate image quality parameters.

  In other words, one or more reports associated with the medical image (s) are analyzed using natural language processing related to the quality of the individual images, for example (one or more) ) Alert information is generated that allows image quality parameters to be generated that can signal the presence of artifacts in the image and that can be used to generate early alerts related to medical devices.

  In this way, a computationally efficient process of reporting for generating image quality parameters can be performed.

  In one example, the classifier module is configured to generate image quality parameters based on image quality issues in at least one portion of the text.

  In other words, text related to imaging issues is analyzed and used to provide information related to the performance of the medical device, thereby monitoring and evaluating the integrity of the medical device. Can provide a good way.

  In this way, an effective and simple way of analyzing text reports is provided for determining the magnitude of image related problems.

  In one example, the classifier module is configured to use a database that includes a plurality of text fragments, and the classifier module converts at least one portion of text into a plurality of texts to generate image quality parameters. Configured to compare with at least one of the fragments.

  This allows image quality parameters to be generated in a simple and efficient manner and allows the learning process to be performed, for example, where the database is extended with text fragments over time.

  In one example, the at least one report includes a plurality of reports, the at least one quality parameter includes a plurality of quality parameters, and the evaluation module generates alert information based on the determined trends in the plurality of quality parameters. Configured to do.

  In this way, when a trend in quality parameters indicates an increasing (eg, severe) problem, this is captured and the problem affects the imaging quality of the medical device or the problem is treated If left untouched, it can be costly to resolve rather than addressing it now before it becomes a problem, allowing corrective action to be taken before showing problems with medical devices can do.

  In one example, the at least one report has been generated by at least one user based on the at least one medical image.

  Thus, taking the example of a medical device that is an X-ray device, the function of the X-ray imaging device can be evaluated by analyzing a report made by a radiologist related to the medical image. This applies equally to MRI devices, CT devices, ultrasound devices and the like.

  In one example, the evaluation module is configured to generate alert information based on a comparison of at least one image quality parameter and a threshold.

  In this way, an efficient means for determining whether there is a problem (or ongoing problem) with a medical device that requires correction is provided.

  In one example, the at least one report includes a plurality of reports, the at least one quality parameter includes a plurality of quality parameters, and the evaluation module is based on the number of the plurality of quality parameters exceeding a threshold value, Configured to generate alert information.

  In this way, a statistical analysis can be provided so that noise in the image does not lead to the generation of alarm information indicative of the problem. Also, if the report indicates that there is a permanent problem in the image collected by the medical device, such as permanent low quality, alert information is generated that allows a customer service ticket to be issued.

According to the second aspect,
-An information providing unit;
An apparatus for assessing medical device quality according to the first aspect;
A system for providing a medical device alert comprising an output unit is provided.

  At least one report is provided from the information providing unit to the input unit. The processing unit is configured to generate alert information based on at least one report provided from the information providing unit. The output unit is configured to output an alarm based on the alarm information.

  In this way, service alerts can be automatically output, eg, to a remote service center, so that maintenance of the medical device can be scheduled if necessary, eg, through a service visit.

According to the third aspect,
b) providing at least one report associated with at least one medical image collected by the medical device, wherein the report is associated with a corresponding medical image;
c) analyzing at least one report and generating at least one image quality parameter, wherein the image quality parameter is associated with the corresponding report;
d) evaluating at least one image quality parameter and generating alert information relating to the medical device, there is provided a method for evaluating medical device quality.

In one example, the method includes:
a) generating at least one report by at least one user based on the at least one medical image;

  According to another aspect, a computer program element control apparatus is provided that, when the computer program element is executed by a processing unit, implements the steps of the method described previously.

  According to another aspect, a computer readable medium storing a computer element as described above is provided.

  Advantageously, the benefits provided by any of the above aspects apply equally to all other aspects, and vice versa.

  The above aspects and examples will be apparent from and elucidated with respect to the embodiments described hereinafter.

  Exemplary embodiments are described below with reference to the following drawings.

FIG. 3 shows a schematic setup of an example of an apparatus for assessing medical device quality. FIG. 2 shows a schematic setup of an example system for providing medical device alerts. FIG. 6 illustrates a method for evaluating medical device quality. FIG. 2 shows a detailed architecture of a system for providing medical alerts along with an adjacent picture archiving communication system (PACS) and adjacent common analyzer tool and early warning system.

  FIG. 1 shows an example of an apparatus 10 for assessing medical device quality. The apparatus 10 includes an input unit 20 and a processing unit 30. The input unit 20 is configured to provide the processing unit 30 with at least one report associated with at least one medical image collected by the medical device, the report being associated with the corresponding medical image. Processing unit 30 is configured to implement a classifier module 40 for analyzing at least one report and generating at least one image quality parameter, the image quality parameter being associated with the corresponding report. The processing unit 30 is also configured to implement an evaluation module 50 for evaluating at least one image quality parameter and generating alert information related to the medical device.

  In one example, each report is associated with a separate medical image. In other words, when a medical image is collected, a change in the quality of the medical image can be determined and correlated with the medical device to implement a corrective action.

  In one example, the report may be associated with one medical image. In other words, the quality varies across the image, is excellent quality on one side of the image, may be acceptable in the middle of the image, and unacceptable on the other side of the image, and this information implements corrective action Can be used for. Thus, here, a report can mean several separate items of information related to an image, or it can mean items of information in a single source related to an image. In other words, a report may mean that different information related to the image is provided, such as comments about quality in different regions of the image, in a single “report” as described above.

  In one example, the classifier module is configured to generate a numerical value for the image quality parameter.

  In this way, an easy-to-analyze quality measure is provided in the form of a numerical value, whereby the image quality associated with one report can be easily and simply associated with another report. To be compared and contrasted.

  In one example, the at least one image quality parameter can be used to signal the presence of at least one artifact in the at least one medical image.

  In one example, the classifier module is configured to distinguish between positive and negative reports. Based on some positive example reports and some negative example reports to the classifier, a positive report, ie a report associated with a good quality image, and a negative report, ie By teaching to distinguish between reports associated with poor quality images, the classifier automatically adapts to the radiologist. That is, if the radiologist usually expresses his thoughts in a negative sense, this is probably present in both the positive and negative examples. As long as the classifier can make a distinction, there are no specific problems when dealing with personal preferences. In this sense, the classifier personalizes the classifier itself according to the radiologist.

  According to an example, the report of the at least one report includes text, and the classifier module 40 is configured to analyze at least one portion of the text to generate image quality parameters.

  In one example, at least one portion of the text is the same as the text of the report.

  In one example, each report of the at least one report includes text that may be different for different reports, and the classifier module may generate different image quality parameters for different reports to generate different image quality parameters. Is configured to analyze at least one portion of the text in.

  In one example, analyzing the at least one portion of text includes parsing at least one portion of the text.

  In one example, the classifier module is configured to generate at least a portion of the text from the text. In one example, the classifier module is configured to remove stop words from the text when generating at least a portion of the text. For example, to extract features such as “image”, “contrast”, “low”, the text string “the image contrast is too low” Can be processed by the classifier module. In one example, the processed features are further subdivided to form n-grams such as “ima”, “mag”, “age”, “con”, “ont”.

  According to one example, the classifier module 40 is configured to generate image quality parameters based on image quality issues in at least one portion of the text.

  In one example, if it is determined that at least a portion of the text is related to an image quality issue, a positive count is associated with the portion.

  In one example, if it is determined that at least a portion of the text is not related to an image quality issue, a negative count is associated with that portion.

  In this way, the classifier module makes a distinction between positive and negative images based on negative comments made by the radiologist in the report associated with the image, and thus the medical device Quality can be evaluated.

  According to one example, the classifier module 40 is configured to use a database 60 that includes a plurality of text fragments. The classifier module 40 is configured to compare at least one portion of text with at least one of the plurality of text fragments to generate image quality parameters.

  In one example, the database contains text fragments known to relate to image quality issues (eg, “low contrast”, “block”, “dot”, “stripes”, “blurred”, “hazy” ”,“ Vague ”,“ Bad ”).

  In one example, the database includes a dictionary. In other words, a dictionary of terms used to describe the problem or problem can be provided, and the text fragments in the report are compared with the terms in the dictionary to generate image quality parameters to quantify the problem. Can be done. In one example, the dictionary includes both positive and negative text fragments to aid the learning process. The positive text fragment is, for example, “high contrast”, “sharp image”, and the like. If only negative text fragments are available, the absence of such text fragments indicates that there are no image quality issues. However, the learning process is improved by providing the ability to process both positive and negative text fragments.

  According to an example, the classifier module 40 is configured to apply a natural language algorithm to the at least one report to generate at least one image quality parameter.

  According to one example, the natural language algorithm includes a learning algorithm. The learning algorithm is configured to generate a value for the image quality parameter based on the at least one training data.

  In one example, the training data is derived from a medical image having an associated report, where problems associated with the image are quantified or verified. In this way, the correlation information forms a data set that the learning algorithm can operate on.

  In one example, the at least one training data is derived from at least one correlation information related to at least one determined quality.

  In one example, the correlation information is derived from a medical image that has an associated report, where problems associated with the image are quantified or verified. In this way, the correlation information forms a data set that the learning algorithm can operate on.

  In one example, the learning algorithm is configured to generate a value for the image quality parameter based on the full text associated with the report. In one example, the learning algorithm is configured to determine radiologist identification information from the associated text in the report. This means that the learning algorithm can distinguish one radiologist from another radiologist based on the language used by different radiologists to identify the radiologist, , Means a certain meaning of “identify”. However, the exemplary language for different radiologists in the database then allows this distinction to provide information related to the probability that the radiologist was a particular radiologist. Which means another meaning of “identify”. In this way, the learning algorithm can determine an overall probability score for a report related to image quality parameters related to the overall negative or global positive meaning of the language used in the report. In this process, the radiologist's identification information can be taken into account. For example, different radiologists may have different views of the same image from the point of view of different uses of negative language, even though the radiologists are equally aware of what the problem is in the image and its severity. Describe. In one example, a learning algorithm can be assisted in this process through a trained professional radiologist and / or medical device specialist reviewing the image and providing baseline reporting information related to the image. This in turn allows comments from different radiologists to be quantified in terms of how the language used by the radiologist relates to the problem.

  In one example, if the report does not contain negative comments, the learning algorithm is configured to determine the probability of negativeness (or positiveness) from all reports, which further develops the learning algorithm, Used to provide information related to medical devices.

  In one example, the entire labeled report can be used as a text fragment so that some text (word, n-gram) that is not directly or clearly related to image quality can also be used in the learning process. This allows personalization tailored to the radiologist, thereby allowing the device to automatically take into account the identity of the user providing the report, and thus in medical device quality. The accuracy of the assessment can be improved because differences in how different users quantify the same or similar imaging problems can be mitigated.

  In this way, it enables the device to learn from annotations by the radiologist with the aim of further improving automatic detection by using a learning algorithm that can be used with a dictionary.

  In one example, the learning algorithm comprises a set of reports associated with associated medical images, and the learning algorithm generates at least one image quality parameter for at least one report that was not previously presented to the device. Scan the set of reports to learn (develop) the (statistical) model used to In one example, the output of the machine learning algorithm is a probability that indicates how much the image can have quality problems. For example, Naive Bayes classifiers calculate positive posterior probabilities based on training examples and using previous probabilities and conditional probabilities. This positive posterior probability plays the above-mentioned role. Similar probabilistic outputs are generally possible with other classifiers.

  As an extension to the learning process, in one example, when the positive posterior probability is close to a negative posterior probability, that is, when both the positive posterior probability and the negative posterior probability are close to 0.5, the device A physician's explicit image quality assessment is determined and added to the positive and negative training sets.

  According to an example, the at least one report includes a plurality of reports and the at least one quality parameter includes a plurality of quality parameters. The evaluation module 50 is configured to generate alert information based on the determined trends in the plurality of quality parameters.

  According to one example, the at least one report is generated by at least one user based on the at least one medical image.

  In one example, a report of at least one report is generated through the processing of verbal input from a radiologist. In this way, the radiologist can verbally describe the image and a report is generated that can be processed by the apparatus to assess medical device quality.

  According to an example, the evaluation module 50 is configured to generate alert information based on a comparison of at least one image quality parameter and a threshold value.

  According to an example, the at least one report includes a plurality of reports and the at least one quality parameter includes a plurality of quality parameters. The evaluation module 50 is configured to generate alarm information based on the number of quality parameters exceeding a threshold value.

  In one example, at least one report is medical data collected by any one or more of the following medical devices: X-ray device, MRI device, PET device, CT device, or ultrasound device. Associated with

  FIG. 2 shows an example of a system 100 for providing medical device alerts. The system 100 includes an information providing unit 110, an apparatus 10 for assessing medical device quality according to any of the examples described with respect to FIG. At least one report is provided from the information providing unit 110 to the input unit 20. The processing unit 30 is configured to generate alarm information based on at least one report provided from the information providing unit 110. The output unit 120 is configured to output an alarm based on the alarm information.

  In one example, the information providing unit is provided in a medical device. In other words, the medical device can collect an image, and a portion of the medical device allows the radiologist to generate at least one report related to the image, or the report is Created and stored in the medical device.

  In one example, the information providing unit is an information storage device.

FIG. 3 shows a method 200 for assessing medical device quality in its basic steps. The method 200 has the following.
In providing step 210, also referred to as step b), at least one report associated with at least one medical image collected by the medical device is provided, and the report is associated with the corresponding medical image.
In an analyzing and generating step 220, also referred to as step c), at least one report is analyzed and at least one image quality parameter is generated, the image quality parameter being associated with the corresponding report.
In an evaluating and generating step 230, also referred to as step d), at least one image quality parameter is evaluated and alarm information relating to the medical device is generated.

  In one example, step c) comprises generating 221 numerical values for the image quality parameters.

  In one example, the report of the at least one report includes text, and step c) comprises analyzing 222 at least one portion of the text to generate an image quality parameter.

  In one example, step c) comprises generating 223 image quality parameters based on image quality issues in at least one portion of the text.

  In one example, step c) comprises using 224 a database that includes a plurality of text fragments, and step c) includes at least one portion of the text to generate a plurality of text quality parameters. It further comprises comparing 225 with at least one of the text fragments.

  In one example, the database includes a dictionary. In one example, the dictionary includes two parts, one part relating to negative words and the second part relating to positive words. This allows the dictionary to be constructed and maintained more easily.

  In one example, step c) comprises applying 226 a natural language processing algorithm to the at least one report to generate at least one image quality parameter.

  In one example, the natural language algorithm includes a learning algorithm and step c) comprises using the learning algorithm 227 to generate a value for the image quality parameter based on the at least one training data.

  In one example, the at least one report includes a plurality of reports, the at least one quality parameter includes a plurality of quality parameters, and step d) includes alert information based on the determined trends in the plurality of quality parameters. Have to generate.

  According to an example, the method comprises in generating step 240, also referred to as step a), that at least one report is generated by at least one user based on the at least one medical image.

  In one example, step d) comprises generating 232 alert information based on the comparison of the at least one image quality parameter and the threshold.

  In one example, the at least one report includes a plurality of reports, the at least one quality parameter includes a plurality of quality parameters, and step d) is based on the number of the plurality of quality parameters exceeding a threshold value. Generating 234 alarm information.

  Next, an apparatus and method for assessing medical device quality and a system for providing medical device alerts are described in more detail in conjunction with FIG.

  FIG. 4 shows a detailed architecture of an environment within which an example system for providing medical device alerts operates, wherein the system includes an example apparatus and method for assessing medical device quality. Take an example. The feature defined by the solid line is a picture archiving communication system (PACS), an example of a system for providing medical device alerts is defined by a dashed line. A system for providing medical device alerts is in communication with a common analyzer tool and an early alert system that a remote service technician has access to. A system for providing medical device alerts is also in communication connection with a PACS radiology information system.

  Referring to FIG. 4, a radiologist views images contained in an image database and possibly uses voice input to write one or more radiology reports to the image database. Have access. Each image has one or several reports associated with it, and some images do not have reports associated with it. The report is information related to the quality of the image, such as describing that part of the image is out of focus, has a low contrast, is poor, or conversely very accurate and good or sufficient Is included. In other words, the report contains information related to the images that the radiologist uses to describe the images and the quality of those images. This input is analyzed using speech recognition and stored as text in a radiology information system (RIS) database. Natural language processing (NLP) systems analyze text radiology reports to extract information related to image quality. The radiologist manually writes the radiology report in text rather than using speech, and the NLP can analyze the radiology report in the same manner. The NLP system can make use of a dictionary specifically designed to help find this information. The image quality assessment system analyzes the image quality data to provide standardized inputs to the common analyzer tool and early warning system that are compatible with the common analyzer tool and early warning system. The learning algorithm learns and thus uses annotations to improve the performance of the NLP system.

  Still referring to FIG. 4, the specific details are as follows.

  The radiologist takes the medical image from the medical image database and creates a text radiology report that is stored in the database of the radiology information system.

  An NLP artifact and IQ problem classifier reads the text radiology report and classifies the text fragment using a learning algorithm previously trained on the annotated report for image quality problems. NLP artifacts and IQ problem classifiers are dictionaries containing text fragments known to be related to image quality problems (eg, “low contrast”, “block”, “dot”, “stripes”, etc.). Can be used.

  The output of the NLP artifact and IQ problem classifier is a set of numerical scores related to image quality. These scores are stored in an image quality database, from which they are evaluated by an image quality assessment system to generate early warnings when there are severe or increasing problems with image quality.

  The remote service technician is notified by the early warning system when the image quality assessment system generates a warning. The remote service technician can also access the image quality assessment system on demand, for example when performing troubleshooting using a common analyzer tool.

Learning Algorithm The learning algorithm trains text fragments related to image quality issues using radiolabel reports that are manually labeled as such (supervised learning). Therefore, text fragments stored in the dictionary are labeled with different positive and negative values. The learning algorithm is a machine learning algorithm such as a neural network, a random forest, or a support vector machine. Other machine learning algorithms can be used.

  For example, assuming a text fragment such as “the image contrast is too low”, the learning algorithm first removes the stopword, then “image”, “contrast” ”,“ Low ”, etc. (typically those terms are further subdivided into n-grams, such as“ ima ”,“ mag ”,“ age ”,“ con ”,“ ont ”, etc. ) The learning algorithm then associates a positive count with these features if the text fragment is labeled as related to an image quality issue, or otherwise associates a negative count with these features. By scanning a large set of annotated fragments, the learning algorithm learns a (statistical) model, which is then not seen before to generate a numerical score related to image quality issues. Can be applied to any text fragment.

Other uses Information from systems for providing medical alerts (which may be considered image quality assessment systems) is also used for other purposes other than as alerts for services, e.g. new in the field It can be used for customer stratification for personalized service level, location / specialty / region comparison in system implementation. Another application of the apparatus and method for assessing medical device quality and systems for providing medical alerts is to detect whether a medical device is being used properly to achieve the best image quality. To provide feedback to technicians or radiology heads, to send specialists to calibrate or better set up medical devices to train technicians using medical devices, or Sometimes even used by service companies. From the embodiments described above, those skilled in the art will clearly understand how these other applications should be implemented.

  In another exemplary embodiment, a computer program or computer program element characterized by being configured to perform the method steps of the method according to one of the preceding embodiments on a suitable system Is provided.

  Accordingly, the computer program element is stored in a computer unit that is also part of the embodiment. The computing unit is configured to perform or induce execution of the method steps described above. Further, the computing unit is configured to operate the apparatus and / or system components described above. The computing unit may be configured to operate automatically and / or execute user instructions. The computer program is loaded into the working memory of the data processor. Accordingly, the data processor is equipped to perform the method according to one of the previous embodiments.

  This exemplary embodiment of the invention covers both computer programs that use the invention from the very beginning and computer programs that, by updating, turn existing programs into programs that use the invention.

  Further, the computer program element can provide all the necessary steps to perform the procedure of the exemplary embodiment of the method described above.

  According to a further exemplary embodiment of the present invention, a computer readable medium is presented, such as a CD-ROM, a USB stick, etc., the computer readable medium storing computer program elements, the computer program elements being described by the preceding section .

  The computer program may be stored and / or distributed on any suitable medium, such as an optical storage medium or solid state medium supplied with or as part of other hardware, but the Internet or other wired or Distributed in other forms, such as via a wireless telecommunications system.

  However, the computer program can also be presented via a network such as the World Wide Web and downloaded from such a network to the working memory of the data processor. According to a further exemplary embodiment of the present invention, a medium is provided for making a computer program element available for download, the computer program element being one of the previously described embodiments of the present invention. Configured to perform the method according to one of the following:

  It should be noted that embodiments of the present invention have been described with respect to different subject matters. In particular, some embodiments have been described with respect to method type claims, while other embodiments have been described with respect to device type claims. However, those skilled in the art, from the above and following description, unless otherwise indicated, in addition to any combination of features belonging to one type of subject matter, any combination between features relating to different subject matter, It can be seen that it is considered disclosed by this application. However, all features can be combined to provide a synergy that goes beyond simple addition of features.

  While the invention has been shown and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive. Not. The invention is not limited to the disclosed embodiments. Other variations to the disclosed embodiments can be understood and implemented by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the dependent claims.

  In the claims, the word “comprising” does not exclude other elements or steps, and the indefinite article “a” or “an” does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that several measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be construed as limiting the scope.

Claims (13)

An apparatus for assessing medical device quality, the apparatus comprising:
An input unit;
A processing unit,
The input unit provides the processing unit with at least one report associated with at least one medical image collected by a medical device, the report associated with the corresponding medical image;
The processing unit implements a classifier module for analyzing the at least one report and generating at least one image quality parameter, wherein the image quality parameter is associated with a corresponding report, the classifier module comprising: Applying a natural language algorithm to the at least one report to generate the at least one image quality parameter, the natural language algorithm including a learning algorithm, the learning algorithm based on at least one training data; Generate values for image quality parameters,
The processing unit implements an evaluation module for evaluating the at least one image quality parameter and generating alert information related to the medical device;
apparatus.   The apparatus of claim 1, wherein a report of the at least one report includes text and the classifier module analyzes at least one portion of the text to generate an image quality parameter.   The apparatus of claim 2, wherein the classifier module generates the image quality parameter based on image quality issues in the at least one portion of the text.   The classifier module uses a database that includes a plurality of text fragments, and the classifier module converts the at least one portion of the text into the plurality of text fragments to generate the image quality parameter. 4. A device according to claim 2 or 3 for comparison with at least one of the following.   The at least one report includes a plurality of reports, the at least one quality parameter includes a plurality of quality parameters, and the evaluation module generates alert information based on the determined trend in the plurality of quality parameters. The device according to claim 1, wherein the device is generated.   The apparatus according to claim 1, wherein the at least one report is generated by at least one user based on the at least one medical image.   The apparatus according to claim 1, wherein the evaluation module generates the alarm information based on a comparison of the at least one image quality parameter and a threshold value.   The at least one report includes a plurality of reports, the at least one quality parameter includes a plurality of quality parameters, and the evaluation module is based on the number of the plurality of quality parameters exceeding the threshold. The apparatus according to claim 7, wherein the alarm information is generated. A system for providing a medical device alert, the system comprising:
An information provision unit;
An apparatus according to any one of claims 1 to 8 for assessing medical device quality;
An output unit,
The at least one report is provided from the information providing unit to the input unit;
The processing unit generates alert information based on the at least one report provided from the information providing unit;
The output unit outputs an alarm based on the alarm information;
system. A method for assessing medical device quality, the method comprising:
b) providing at least one report associated with at least one medical image collected by the medical device, wherein the report is associated with a corresponding medical image;
c) applying a natural language processing algorithm including a learning algorithm to the at least one report and using the learning algorithm to generate a value for an image quality parameter based on the at least one training data. Analyzing the at least one report and generating at least one image quality parameter, wherein the image quality parameter is associated with a corresponding report; and
d) evaluating the at least one image quality parameter and generating alert information related to the medical device. The method comprises
11. The method of claim 10, comprising the step of: a) generating the at least one report by at least one user based on the at least one medical image.   A computer program for controlling the apparatus according to any one of claims 1 to 9, wherein the computer program performs the method according to claim 10 or 11 when executed by a processor.   A computer-readable medium storing the computer program according to claim 12.

Images